Higher normalized concentrations of tetracycline resistance found in ballast and harbor water compared to ocean water

Higher normalized concentrations of tetracycline resistance found in ballast and harbor water compared to ocean water

Marine Pollution Bulletin 151 (2020) 110796 Contents lists available at ScienceDirect Marine Pollution Bulletin journal homepage: www.elsevier.com/l...

330KB Sizes 0 Downloads 21 Views

Marine Pollution Bulletin 151 (2020) 110796

Contents lists available at ScienceDirect

Marine Pollution Bulletin journal homepage: www.elsevier.com/locate/marpolbul

Higher normalized concentrations of tetracycline resistance found in ballast and harbor water compared to ocean water William A. Gerhard, Claudia K. Gunsch

T



Department of Civil and Environmental Engineering, Duke University, Durham, NC 27708, United States

A B S T R A C T

Although ballast water is a known vector for the global transport of microorganisms, the Ballast Water Management Convention only sets limits for indicator organisms and does not consider antibiotic resistance genes (ARGs). Herein, we examined the concentration of indicator organisms and prevalence of three ARGs (sul1, tet (M), and vanA) in a total of 53 ballast, 21 harbor, and 8 ocean samples collected in Singapore, China, South Africa, and California. E. coli was found in significantly higher concentrations in ballast samples obtained in Singapore and China compared to South Africa (Singapore, p = 0.040) and California (Singapore, p < 0.001; China, p = 0.038). Harbor samples from China had significantly higher concentrations of E. coli than Singapore (p = 0.049) and California (p = 0.001). When compared to ocean samples, there were significantly higher concentrations of normalized tet(M) in ballast samples from California (p = 0.011) and Singapore (p = 0.019) and in harbor samples from California (p = 0.018), Singapore (p = 0.010), and South Africa (p = 0.008). These findings suggest that microbial loads significantly differ among ports. Furthermore, certain ARGs are enriched in ballast and harbor waters when compared to ocean water, which suggests that ballast waters have the potential to either transport higher concentrations of certain ARGs or that ballast tank conditions may exert selective pressure for some ARGs.

1. Introduction Since 2013, over 280 million metric tons of ballast water have been discharged annually into United States ports (Gerhard and Gunsch, 2018). Ballast water is a known vector for the global transport of microorganisms (Drake et al., 2005, 2002; Mimura et al., 2005; Ruiz et al., 2000). The spread of these organisms can have far-reaching effects on humans and the environment, which was exemplified by a ballast-associated outbreak of Vibrio cholerae in Peru during the 1990s that killed > 10,000 people (McCarthy and Khambaty, 1994). Research to examine the microbial characteristics of ballast waters is especially timely because sea ice is expected to continue receding during the 21st century, which will open additional reaches of the ocean to maritime shipping and its possible impacts on human and environmental health (Buixadé Farré et al., 2014; Holbech and Pederson, 2018; Stephenson et al., 2013). The International Maritime Organization (IMO) drafted the Ballast Water Management (BWM) Convention in 2004 to address some of the concerns regarding harmful aquatic organisms (Gollasch et al., 2007). The BWM Convention includes two management standards that encourage protection of aquatic environments from invasive or pathogenic species. Specifically, Standard D-2 established enforceable compliance standards for microbial ballast water quality: E. coli (< 250 viable organisms per 100 mL) and Enterococcus spp (< 100 viable organisms per 100 mL) (Lloyd's Register, 2010). The BWM Convention entered into force in September 2017 and has been met with an



increase in the utilization of ballast water treatment systems; however, the literature that examines indicator organism distribution in ballast water remains relatively sparse (Gerhard et al., 2019). Perhaps the most ominous threat to human and environmental health presented by ballast water, and one that is notably absent from mention in the BWM Convention, is the role of ballast water as a possible medium for the global transport of antibiotic resistance. Antibiotic resistant strains of Vibrio have been found in ballast water arriving to major hub ports, highlighting the possibility that ballast may be inadvertently transporting antibiotic resistance globally (Ng et al., 2018; Ruiz et al., 2000). Identifying whether there are significantly different concentrations of antibiotic resistance genes (ARGs) in ballast water tanks when compared to harbor and open ocean samples could provide further insight to the role of ballast in global dissemination of ARGs. In this study, we examine indicator organisms proposed by the IMO (E. coli and Enterococcus spp) in addition to total coliforms in order to assess differences among locations and sample types. In addition, we explore the prevalence of antibiotic resistance across sample types and locations by enumerating the ARGs via qPCR. We selected three ARGs [sul1 (sulfonamide), tet(M) (tetracycline), and vanA (vancomycin)] that range from nearly ubiquitous (sul1) to rarely occurring in natural environments (vanA) (Chang et al., 2003; Ng et al., 2015; Volkmann et al., 2004). Significantly different normalized concentrations of ARGs among sample types or locations could merit further inspection to identify the best practices to prevent ARG transmission via high

Corresponding author at: Duke University, Department of Civil and Environmental Engineering, 121 Hudson Hall, Durham, NC 27708-0287, United States. E-mail address: [email protected] (C.K. Gunsch).

https://doi.org/10.1016/j.marpolbul.2019.110796 Received 10 May 2019; Received in revised form 1 December 2019; Accepted 1 December 2019 Available online 29 January 2020 0025-326X/ © 2019 Elsevier Ltd. All rights reserved.

Marine Pollution Bulletin 151 (2020) 110796

W.A. Gerhard and C.K. Gunsch

The copies of bacterial 16S rRNA gene were determined using the standard protocol for 16S rRNA gene amplification described by the Earth Microbiome Project (Apprill et al., 2015; Caporaso et al., 2011; Parada et al., 2016). The primers used in this study are: 515F (Parada) 5′ – GTGYCAGCMGCCGCGGTAA – 3′; and 806R (Apprill) 5′ – GGACTACNVGGGTWTCTAAT – 3′. Cycling conditions were: 1) initiation at 94 °C for 3 min; 2) 35 cycles at 94 °C for 45 s, 50 °C for 60 s, and 72 °C for 90 s; and 3) 72 °C for 10 min. A melt curve was performed from 70 to 90 °C with steps of 0.4 °C every 10 s. Three different ARGs were tested in this analysis: 1) sul1 – encoding sulfonamide resistance via dihydropteroate synthase encoding genes (Hu et al., 2008; Moura et al., 2007); 2) tet(M) – encoding tetracycline resistance via ribosomal protection protein (Aminov et al., 2004); and 3) vanA – encoding vancomycin resistance via vancomycin resistance protein (Schwartz et al., 2003). The tet(M) and vanA ARGs were examined using a multiplexed qPCR approach, and the sul1 assay was performed separately. The primers and probes used to in this study were described previously: sul1 (Czekalski et al., 2012); tet(M) (Peak et al., 2007); and vanA (Volkmann et al., 2004) (Supplementary Table 1). The cycling conditions used for these assays were standard TaqMan cycling protocol: 1) initiation at 50 °C for 2 min and 94 °C for 10 min; and 2) 40 cycles at 95 °C for 15 s and at 60 °C for 60 s.

prevalence matrices or locations. 2. Materials and methods 2.1. Sample collection A total of 82 ballast, harbor, and ocean water samples were collected from four ports over a two-year period from September 2015 to August 2017. The ports included in this study are (ballast samples, harbor samples): Los Angeles/Long Beach, California (34, 8); Durban, South Africa (4, 4); Shanghai, China (7, 5); and Singapore (8, 4). For simplicity, these locations will be referred to as California, South Africa, China, and Singapore throughout this paper. There were eight open ocean samples collected on a sail boat during May 2016 from Singapore and Jakarta, Indonesia (Table S1). The sampling procedures for ballast, harbor, and ocean water samples were similar. Three distinct 1.2 L samples were collected approximately 1 m below the water surface using a 1.2 L Kemmerer sampler (Wildco, Yulee, Florida). Ballast samples were collected through an open manhole, harbor samples were collected by lowering the sampler from the edge of the dock, and ocean samples were collected by lowering the sampler from the side of the sailboat. All samples were stored in glass bottles on ice for transport to the laboratory.

2.4. Data analysis

2.2. Indicator organism enumeration

Culture data was logarithmically transformed for data analysis. All culture data below the lower limit of detection (LLOD) with the IDEXX kit (i.e. 1 MPN per 100 mL) was recorded as ( LLOD ) as has been pre2 viously described for IDEXX MPN data (Gerhard et al., 2017). Samples above the upper limit of quantification (i.e. 2419.6 MPN per 100 mL) were assigned the value of this upper limit. It is important to note that the 1:10 dilution resulted in a quantifiable range of 10–24,196 MPN per 100 mL. The arithmetic mean of the logarithmically transformed data (i.e. geometric mean) for each sample type (i.e. ballast and harbor) was tested for significance using an unequal variances two sample t-test. All means were reported as an untransformed version of the geometric mean. A one-way ANOVA was performed to examine whether there were significant differences among sample locations. In the case of significance, Tukey's post-hoc test was used to determine which independent variables were significantly different than the others. Molecular qPCR data for each sample was determined by averaging the cycle threshold value for each sample and calculating the value using the standard curve. All qPCR data below the lower limit of quantification (LLOQ) for a given standard curve was transformed to a value of 1. The ARG concentration was normalized to 16S rRNA gene concentration for each sample. The resulting ratio was then logarithmically transformed. This process has been described previously (Li et al., 2015). These log ratios were compared across sample type and sample location using one-way ANOVA and Tukey's post-hoc test as described for the culture data.

Culture-based analyses were performed on all distinct ballast and harbor water samples. Lack of incubation capability while at sea prevented culture-based analysis of ocean samples. Indicator organisms were cultured using the IDEXX Colilert-18 (total coliforms, E. coli) and Enterolert (intestinal enterococci) assays according to the manufacturer's protocol with a 1:10 dilution for all samples to decrease the effects of salinity. These assays are substrate tests that can be used to detect enzymes specific to the targeted indicator organisms. The United States Environmental Protection Agency (USEPA) has approved methods for water quality analysis using these assays (Microbial Contaminants Method 9223, 2005). 2.3. Molecular analyses The sample preparation techniques for molecular analyses used in this study have been previously described (Gerhard and Gunsch, 2019). Briefly, 1 L of each sample was filtered through a 0.45 μm polycarbonate filter within 12 h of collection. The filter paper was stored at −20 °C prior to transport to Duke University for DNA extraction and analysis. DNA extraction was performed using the MoBio PowerSoil® DNA Isolation Kit (Carlsbad, CA USA) according to the manufacturer's protocol. All quantitative real-time PCR (qPCR) analyses were performed at Duke University using a Bio-Rad CFX96 Touch Real-Time PCR Detection System (Hercules, California, USA). Standard curves and no template controls (NTCs) were generated for every run of each assay. All samples were analyzed in triplicate. The following sequences were used in this study for standard curves, assay (accession number): 1) 16S rRNA gene (NR_157609.1); 2) sul1 (KY887591.1); 3) tet(M) (JINK01000010); and 4) vanA (KX810026). Gene fragments (gBlocks) with known concentrations were used to generate all standard curves (Integrated DNA Technologies; Skokie, Illinois, USA). All assays were performed in 25 μL volume with the following mixture components: 1) 13 μL of PCR-grade water; 2) 10 μL of master mix; 3) 1 μL of primer or primer/probe mix – described in each section; and 4) 1 μL of template DNA. The 16S rRNA gene assay used SYBR chemistry with iTaq™ Universal SYBR® Green Supermix (Bio-Rad Laboratories, Inc.; Hercules, California, USA). All ARG assays used TaqMan™ chemistry with iTaq™ Universal Probes Supermix (Bio-Rad Laboratories, Inc.; Hercules, California, USA).

3. Results and discussion 3.1. Culture analysis Culture-based comparison of sample types was performed to identify significant differences between ballast and harbor water samples segmented by location (Table 1). When aggregating all locations, significant differences were observed between ballast and harbor water samples regarding total coliforms (p = 0.001) and E. coli (p = 0.007). The reported geometric mean and 95% confidence interval of aggregated samples was lower in ballast (total coliforms, 47 [23–98] MPN per 100 mL; E. coli, 9.9 [7.4–13] MPN per 100 mL) than harbor water (total coliforms, 628 [185–2133] MPN per 100 mL; E. coli, 36 [15–85] MPN per 100 mL). The specific locations often did not have sufficient 2

Marine Pollution Bulletin 151 (2020) 110796

W.A. Gerhard and C.K. Gunsch

Table 1 Comparison of indicator organism concentrations between ballast and harbor water samples. All values are most probable number per 100 mL reported as geometric mean with 95% confidence interval. Location

California (n = 34, 8)1 China (n = 7, 5)1 Singapore (n = 8, 4)1 South Africa (n = 4, 4)1 All (n = 53, 21)1 1 2 ⁎

Total coliforms

E. coli

Enterococcus spp

Ballast

Harbor

p-Value

Ballast

Harbor

p-Value

Ballast

Harbor

p-Value

11 (7.1–18) 1828 (215–15,576) 858 (225–3264) 44 (0.7–2782) 47 (23–98)

107 (17–680) 10,662 (5031–22,598) 268 (81–888) 2628 (0.2–3.7 × 107) 628 (185–2133)

0.025⁎

6.8 (6.5-7.0) 18 (4.3–74) 34 (8.2–138) All below LLOD2

8.4 (4.9–15) 327 (26–4077) 29 (8.5–104)

0.370

0.184

7.7 (6.4–9.3) 246 (108–559) 14 (2.4–87) 88 (5.7–1387) 26 (12–56)

0.272

59 (0.6–5551) 36 (15–85)

7.0 (6.6–7.4) 23 (7.5-73) 132 (8.1–2154) All below LLOD2

0.095 0.130 0.198 0.001⁎

9.9 (7.4-13)



0.030

0.837

0.007⁎

13 (8.1-21)

0.002⁎ 0.130 0.058 0.117

Reported as n = ballast, harbor. All values below the lower limit of detection (10 MPN per 100 mL). Significant at alpha = 0.05.

Fig. 1. (A) Log10 transformed concentration of indicator organisms grouped by sample type. (***) indicates a significant difference (p < 0.05) among locations at the endpoints of each line. (B) Log10 transformed ratio of antibiotic resistance gene copy numbers to 16S rRNA gene copy numbers grouped by sample type. (***) indicates a significant difference (p < 0.05) among locations at the endpoints of each line. (**) indicates a significant difference with the ocean samples.

Enterococcus spp concentration among harbor samples existed among China and California (p < 0.001), South Africa and California (p = 0.004), Singapore and China (p < 0.001), and South Africa and Singapore (p = 0.048). Differences in indicator organism concentration among harbors highlight the differential potential to move microbes around the world in ballast water. This finding has particular relevance to vessels attempting to comply with IMO D-2 Standards. Loading ballast in harbors with high indicator organism concentrations, Shanghai and Durban in this study, should be avoided when possible to reduce the likelihood of failure in the event of compliance testing.

sample size to generate a significant p-value despite patterns that generally reflected the aggregated results in California, China, and South Africa with higher concentrations in harbor than ballast water (Fig. 1A). Notably, Singapore samples had a higher concentration of indicator organisms in ballast than harbor. This observation may be because ballast water samples collected in Singapore were often loaded while the vessel was in Chinese harbors – Shanghai harbor samples had the highest geometric mean of indicator organisms compared to all other sites by at least an order of magnitude in every tested category. Furthermore, there were significant differences among locations when segmented by sample type (Fig. 1A). Generally, harbor water samples from Singapore and California had lower concentrations of indicator organisms than harbor water samples from South Africa and China, and the concentration of indicator organisms in ballast water samples were generally higher in Singapore and China than South Africa and California. However, these differences were not always significant. Significant differences in total coliform concentration among ballast samples existed among China and California (p < 0.001), Singapore and California (p < 0.001), South Africa and China (p = 0.006), and South Africa and Singapore (p = 0.035). Significant differences in E. coli concentration among ballast samples existed among China and California (p = 0.038), Singapore and California (p < 0.001), and South Africa and Singapore (p = 0.040). Significant differences in Enterococcus spp concentration among ballast samples existed among Singapore and California (p < 0.001) and South Africa and Singapore (p = 0.013). Significant differences in total coliform concentration among harbor samples existed in China and California (p = 0.005) and Singapore and China (p = 0.046). Significant differences in E. coli concentration among harbor samples existed among China and California (p = 0.001) and Singapore and China (p = 0.049). Significant differences in

3.2. Molecular analysis ARG copy numbers were quantified by qPCR and normalized to 16S rRNA gene copies in each sample. For each assay, the lowest standard curve dilution with consistent performance was (in terms of the original sample concentration): 16S rRNA gene (10 copies per mL); sul1 (1 copy per mL); tet(M) (10 copies per mL); and vanA (10 copies per mL). The percentage of samples above the LLOQ for each qPCR assay was: 16S rRNA gene (100%); sul1 (96.4%); tet(M) (74.4%); and vanA (0%). VanA is not reported in any of the figures or tables because this ARG was not detected in any samples. All copy concentration data can be found in the supplementary material (Table S2). Comparison of normalized sul1 in ballast, harbor, and ocean water across all locations revealed that there were not many significant differences among sample types (Table 2). The only significantly different value was a higher log ratio in Chinese harbor water than Chinese ballast (p = 0.033) and open ocean water (p = 0.005) (Fig. 1B). This finding may provide evidence that factors influencing sul1 selection could be similar across locations and sample types in this study. There were significantly lower normalized tet(M) values in ocean samples 3

Marine Pollution Bulletin 151 (2020) 110796

W.A. Gerhard and C.K. Gunsch

Table 2 Comparison of log ratio of antibiotic resistance gene copies to 16S rRNA gene copies. All values are average log ratio with a 95% confidence interval. Location

sul1

tet(M) 1

Ballast

Harbor

Ocean

California (n = 34, 8)3

−3.4 (−3.7 to −3.2)

−3.8 (−4.9 to −2.7)

−3.5 (−4.5 to −2.5)

China (n = 7, 5)3

−3.0 (−3.5 to −2.6)

−1.5 (−2.7 to −0.4)

−3.5 (−4.5 to −2.5)

Singapore (n = 8, 4)3

−3.5 (−4.6 to −2.5)

−4.4 (−6.1 to −2.7)

−3.5 (−4.5 to −2.5)

South Africa (n = 4, 4)3

−2.6 (−4.3 to −0.9)

−2.1 (−3.9 to −0.3)

−3.5 (−4.5 to −2.5)

All

−3.3 (−3.6 to −3.1)

−3.2 (−3.9 to −2.4)

−3.5 (−4.5 to −2.5)

(n = 53, 21, 8)4

1 2 3 4 ⁎

p-Value

2

BH – 0.529 BO – 0.988 HO – 0.744 BH – 0.033⁎ BO – 0.661 HO – 0.005⁎ BH – 0.497 BO – 0.993 HO – 0.442 BH – 0.844 BO – 0.435 HO – 0.217 BH – 0.829 BO – 0.937 HO – 0.778

Ballast

Harbor

Ocean1

p-Value2

−4.3 (−4.7 to −3.9)

−4.0 (−4.5 to −3.5)

−5.5 (−6.4 to −4.7)

−5.6 (−7.3 to −4.0)

−3.9 (−7.1 to −0.7)

−5.5 (−6.4 to −4.7)

−3.6 (−5.1 to −2.2)

−3.2 (−3.5 to −2.8)

−5.5 (−6.4 to −4.7)

−4.7 (−6.8 to −2.6)

−2.9 (−4.7 to −1.1)

−5.5 (−6.4 to −4.7)

−4.4 (−4.8 to −4.0)

−3.6 (−4.2 to −3.0)

−5.5 (−6.4 to −4.7)

BH – 0.835 BO – 0.011⁎ HO – 0.018⁎ BH – 0.244 BO – 0.993 HO – 0.267 BH – 0.786 BO – 0.019⁎ HO – 0.010⁎ BH – 0.109 BO – 0.426 HO – 0.008⁎ BH – 0.087 BO – 0.073 HO – 0.003⁎

All ocean samples are the same 8 samples collected in the South China Sea. Comparisons listed in order of ballast-harbor (BH), ballast-ocean (BO), and harbor-ocean (HO). Reported as n = ballast, harbor. Reported as n = ballast, harbor, ocean. Significant at alpha = 0.05.

under grants no. DGE 1545220, no. OISE 1243433, no. OISE 1713717, and no. OISE 1614161. Further support was provided through a Graduate Student Training Enhancement Grant from Duke University Interdisciplinary Studies, and a Graduate Research and Training Award from the Duke University Center for International and Global Studies.

when compared to harbor samples in Singapore, California, South Africa, and when aggregating all locations (Fig. 1B, Table 2). These differences could be related to proximity to anthropogenic impacts that occur around harbor areas such as industrial runoff or wastewater discharge, which may create selective pressure for the tet(M) ARG. Finally, normalized tet(M) values in the ocean samples were significantly lower than the ballast samples collected in Singapore and California (Fig. 1B). A higher normalized value for the tet(M) ARG in ballast and harbor waters when compared to ocean water highlights the potential of ballast to serve as a vector for the global dispersion of ARGs. Additional research is necessary to explore the relationship between ballast, harbor, and ocean waters throughout the voyage that may contribute to the observed differences. The vanA gene was not found in any samples, which is an encouraging sign regarding the transmission of resistance to this last line antibiotic in ballast water. The data presented herein suggest that additional research into the possible risks of ballast-associated ARG transfer is warranted to effectively protect human and environmental health. This study utilizes a relatively small sample size and geographic range of open ocean samples, so future research to compare other open ocean data sets to the harbor and ballast data reported herein would prove particularly insightful. In addition, further research is required to understand the effect of different ballast water treatment systems on prevalence of ARGs and indicator bacteria.

Appendix A. Supplementary data Supplementary data to this article can be found online at https:// doi.org/10.1016/j.marpolbul.2019.110796. References Aminov, R.I., Chee-Sanford, J.C., Garrigues, N., Mehboob, A., Mackie, R.I., 2004. Detection of tetracycline resistance genes by PCR methods. In: Spencer, J.F.T., Ragout de Spencer, A.L. (Eds.), Public Health Microbiology. Humana Press, New Jersey, pp. 003–014. https://doi.org/10.1385/1-59259-766-1:003. Apprill, A., Mcnally, S., Parsons, R., Weber, L., 2015. Minor revision to V4 region SSU rRNA 806R gene primer greatly increases detection of SAR11 bacterioplankton. Aquat. Microb. Ecol. https://doi.org/10.3354/ame01753. Buixadé Farré, A., Stephenson, S.R., Chen, L., Czub, M., Dai, Y., Demchev, D., Efimov, Y., Graczyk, P., Grythe, H., Keil, K., Kivekäs, N., Kumar, N., Liu, N., Matelenok, I., Myksvoll, M., O’Leary, D., Olsen, J., Pavithran., A.P.S., Petersen, E., Raspotnik, A., Ryzhov, I., Solski, J., Suo, L., Troein, C., Valeeva, V., van Rijckevorsel, J., Wighting, J., 2014. Commercial Arctic shipping through the Northeast Passage: routes, resources, governance, technology, and infrastructure. Polar Geogr. https://doi.org/10. 1080/1088937X.2014.965769. Caporaso, J.G., Lauber, C.L., Walters, W.A., Berg-Lyons, D., Lozupone, C.A., Turnbaugh, P.J., Fierer, N., Knight, R., 2011. Global patterns of 16S rRNA diversity at a depth of millions of sequences per sample. Proc. Natl. Acad. Sci. https://doi.org/10.1073/ pnas.1000080107. Chang, S., Sievert, D.M., Hageman, J.C., Boulton, M.L., Tenover, F.C., Downes, F.P., Shah, S., Rudrik, J.T., Pupp, G.R., Brown, W.J., Cardo, D., Fridkin, S.K., 2003. Infection with vancomycin-resistant Staphylococcus aureus containing the vanA resistance gene. N. Engl. J. Med. 348, 1342–1347. https://doi.org/10.1056/NEJMoa025025. Czekalski, N., Berthold, T., Caucci, S., Egli, A., Bürgmann, H., 2012. Increased levels of multiresistant bacteria and resistance genes after wastewater treatment and their dissemination into Lake Geneva, Switzerland. Front. Microbiol. https://doi.org/10. 3389/fmicb.2012.00106. Drake, L.A., Ruiz, G.M., Galil, B.S., Mullady, T.L., Friedmann, D.O., Dobbs, F.C., 2002. Microbial ecology of ballast water during a transoceanic voyage and the effects of open-ocean exchange. Mar. Ecol. Prog. Ser. 233, 13–20. https://doi.org/10.3354/ meps233013. Drake, L.A., Meyer, A.E., Forsberg, R.L., Baier, R.E., Doblin, M.A., Heinemann, S., Johnson, W.P., Koch, M., Rublee, P.A., Dobbs, F.C., 2005. Potential invasion of microorganisms and pathogens via “interior hull fouling”: biofilms inside ballast water tanks. Biol. Invasions 7, 969–982. https://doi.org/10.1007/s10530-004-3001-8. Gerhard, W.A., Gunsch, C.K., 2018. Analyzing trends in ballasting behavior of vessels arriving to the United States from 2004 to 2017. Mar. Pollut. Bull. 135, 525–533. https://doi.org/10.1016/j.marpolbul.2018.07.001.

CRediT authorship contribution statement William A. Gerhard: Conceptualization, Methodology, Investigation, Validation, Data curation, Funding acquisition, Visualization, Writing - original draft.Claudia K. Gunsch: Supervision, Funding acquisition, Writing - review & editing. Declaration of competing interest The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper. Acknowledgements This work was supported by the National Science Foundation (NSF) 4

Marine Pollution Bulletin 151 (2020) 110796

W.A. Gerhard and C.K. Gunsch

Moura, A., Henriques, I., Ribeiro, R., Correia, A., 2007. Prevalence and characterization of integrons from bacteria isolated from a slaughterhouse wastewater treatment plant. J. Antimicrob. Chemother. 60, 1243–1250. https://doi.org/10.1093/jac/ dkm340. Ng, C., Le, T.H., Goh, S.G., Liang, L., Kim, Y., Rose, J.B., Yew-Hoong, K.G., 2015. A comparison of microbial water quality and diversity for ballast and tropical harbor waters. PLoS One. https://doi.org/10.1371/journal.pone.0143123. Ng, C., Goh, S.G., Saeidi, N., Gerhard, W.A., Gunsch, C.K., Gin, K.Y.H., 2018. Occurrence of Vibrio species, beta-lactam resistant Vibrio species, and indicator bacteria in ballast and port waters of a tropical harbor. Sci. Total Environ. 610–611, 651–656. https://doi.org/10.1016/j.scitotenv.2017.08.099. Parada, A.E., Needham, D.M., Fuhrman, J.A., 2016. Every base matters: assessing small subunit rRNA primers for marine microbiomes with mock communities, time series and global field samples. Environ. Microbiol. https://doi.org/10.1111/1462-2920. 13023. Peak, N., Knapp, C.W., Yang, R.K., Hanfelt, M.M., Smith, M.S., Aga, D.S., Graham, D.W., 2007. Abundance of six tetracycline resistance genes in wastewater lagoons at cattle feedlots with different antibiotic use strategies. Environ. Microbiol. https://doi.org/ 10.1111/j.1462-2920.2006.01123.x. Ruiz, G.M., Rawlings, T.K., Dobbs, F.C., Drake, L.A., Mullady, T., Huq, A., Colwell, R.R., 2000. Global spread of microorganisms by ships. Nature 408, 49–50. https://doi.org/ 10.1038/35040695. Schwartz, T., Kohnen, W., Jansen, B., Obst, U., 2003. Detection of antibiotic-resistant bacteria and their resistance genes in wastewater, surface water, and drinking water biofilms. FEMS Microbiol. Ecol. 43, 325–335. https://doi.org/10.1111/j.1574-6941. 2003.tb01073.x. Stephenson, S.R., Smith, L.C., Brigham, L.W., Agnew, J.A., 2013. Projected 21st-century changes to Arctic marine access. Clim. Chang. 118, 885–899. https://doi.org/10. 1007/s10584-012-0685-0. Volkmann, H., Schwartz, T., Bischoff, P., Kirchen, S., Obst, U., 2004. Detection of clinically relevant antibiotic-resistance genes in municipal wastewater using real-time PCR (TaqMan). J. Microbiol. Methods. https://doi.org/10.1016/j.mimet.2003.10. 014.

Gerhard, W.A., Gunsch, C.K., 2019. Metabarcoding and machine learning analysis of environmental DNA in ballast water arriving to hub ports. Environ. Int. 124, 312–319. https://doi.org/10.1016/J.ENVINT.2018.12.038. Gerhard, W.A., Choi, W.S., Houck, K.M., Stewart, J.R., 2017. Water quality at points-ofuse in the Galapagos Islands. Int. J. Hyg. Environ. Health 220, 485–493. https://doi. org/10.1016/j.ijheh.2017.01.010. Gerhard, W.A., Lundgreen, K., Drillet, G., Baumler, R., Holbech, H., Gunsch, C.K., 2019. Installation and use of ballast water treatment systems - implications for compliance and enforcement. Ocean Coast. Manag. 104907. https://doi.org/10.1016/j. ocecoaman.2019.104907. Gollasch, S., David, M., Voigt, M., Dragsund, E., Hewitt, C., Fukuyo, Y., 2007. Critical review of the IMO international convention on the management of ships’ ballast water and sediments. Harmful Algae 6, 585–600. https://doi.org/10.1016/j.hal. 2006.12.009. Holbech, H., Pederson, K.L., 2018. Ballast water and invasive species in the Arctic. In: Vestergaard, N., Kaiser, B.A., Fernandez, L., Larsen, J.N. (Eds.), Arctic Marine Resource Governance and Development, pp. 115–137. Hu, J., Shi, J., Chang, H., Li, D., Yang, M., Kamagata, Y., 2008. Phenotyping and genotyping of antibiotic-resistant Escherichia coli isolated from a natural river basin. Environ. Sci. Technol. 42, 3415–3420. https://doi.org/10.1021/es7026746. Li, B., Yang, Y., Ma, L., Ju, F., Guo, F., Tiedje, J.M., Zhang, T., 2015. Metagenomic and network analysis reveal wide distribution and co-occurrence of environmental antibiotic resistance genes. ISME J 9, 2490–2502. https://doi.org/10.1038/ismej. 2015.59. Lloyd'’s Register, 2010. Ballast Water Treatment Technology: Current Status, February 2010. McCarthy, S.A., Khambaty, F.M., 1994. International dissemination of epidemic Vibrio cholerae by cargo ship ballast and other nonpotable waters. Appl. Environ. Microbiol. 60, 2597–2601. Microbial Contaminants Method 9223, 2005. Standard Methods for the Examination of Water and Wastewater. Standard Methods. Mimura, H., Katakura, R., Ishida, H., 2005. Changes of microbial populations in a ship’s ballast water and sediments on a voyage from Japan to Qatar. Mar. Pollut. Bull. 50, 751–757. https://doi.org/10.1016/j.marpolbul.2005.02.006.

5